阅读说明：本技术 一种蓝牙耳机检测疲劳的方法 (Method for detecting fatigue of Bluetooth headset ) 是由 屈军锁 乔宁 于 2019-10-28 设计创作，主要内容包括：本发明提供一种蓝牙耳机检测疲劳的方法,蓝牙耳机同正常耳机一样可接打电话、听音乐,在此基础上集成疲劳驾驶检测模块,获取驾驶员驾驶过程中头部的加速度、角速度信息,通过对驾驶员头部姿态的估计,同时结合发生低头动作的冲击时长,设计两级阈值判定,当检测到疲劳打盹引起头部低头状态时,则认定驾驶员处于疲劳驾驶状态,此时触发报警器声光报警。检测方法方便、高效、简单,可有效的减少交通事故的发生。(The invention provides a method for detecting fatigue by a Bluetooth headset, wherein the Bluetooth headset can make a call and listen to music as a normal headset, a fatigue driving detection module is integrated on the basis to acquire the acceleration and angular velocity information of the head of a driver in the driving process, two-stage threshold value judgment is designed by estimating the head posture of the driver and combining the impact duration of head lowering action, when the head lowering state caused by fatigue napping is detected, the driver is determined to be in the fatigue driving state, and an audible and visual alarm is triggered at the moment. The detection method is convenient, efficient and simple, and can effectively reduce the occurrence of traffic accidents.)

1. A method for detecting fatigue of a Bluetooth headset is characterized by comprising the following steps: the system mainly comprises an audio encoding and decoding module, a fatigue detection device, a network module, a power supply module and an acousto-optic alarm; the Bluetooth headset can make and listen to the phone as a normal headset, a fatigue driving detection module is integrated on the basis to acquire the acceleration and angular velocity information of the head of a driver in the driving process, filtering processing is carried out before the acquired data are used to reduce the influence of noise on the data, meanwhile, two-stage threshold value judgment is designed by combining impact duration, when the head is in a low head state caused by fatigue napping, the driver is determined to be in the fatigue driving state, and an audible and visual alarm is triggered at the moment.

2. A method for detecting fatigue of a Bluetooth headset is characterized by comprising the following steps:

after the Bluetooth headset is started, all sensor modules in the Bluetooth headset are initialized.

Acquiring triaxial acceleration data and triaxial angular velocity data of the head posture of the human body through a sensor;

filtering is needed to reduce the influence of noise before the original data are used, and a relative optimal value at the current moment is obtained;

setting the acceleration intensity value and the impact duration as a primary threshold, and if the acceleration intensity value and the impact duration exceed the primary threshold, judging the suspected fatigue driving state;

after the suspected fatigue state is judged, setting the angular velocity strength value as an earphone threshold, and judging the fatigue driving state if the angular velocity strength value exceeds a secondary threshold condition;

and after the fatigue driving state is judged, the audible and visual alarm gives an alarm to remind a driver of driving safely.

3. The method of claim 1, wherein the method comprises the following steps: the fatigue driving detection module is used for estimating the head posture condition, and triggering the audible and visual alarm to give an alarm if the data value meets the set threshold value.

4. The method of claim 1, wherein the method comprises the following steps: judging whether fatigue driving is performed according to the double thresholds; and if the first threshold condition of the double thresholds is met, the suspected fatigue driving is judged, and if the second threshold condition is met, the fatigue driving is judged, and the audible and visual alarm gives an alarm.

5. The method of claim 1, wherein the method comprises the following steps: the fatigue state refers to unsafe driving behavior in which the driver lowers his head sharply due to dozing off, causing rapid changes in acceleration and angular velocity in the three-dimensional direction.

6. The method of claim 5, wherein the method comprises the following steps: the acceleration and angular velocity information in the three-dimensional direction is set to establish a space coordinate system through the three-dimensional direction in the actual driving process of the driver, and the acceleration and angular velocity information of the head of the driver in the driving process is acquired by using the sensor module, so that the threshold value is judged.

7. The method of claim 2, wherein the method comprises the following steps: the sensor module is initialized, on one hand, the common functions of the earphone (including call receiving and calling, music listening and the like) can be normally used, and on the other hand, the fatigue driving detection module is initialized.

8. The method of claim 2, wherein the method comprises the following steps: and the filtering processing is to use Kalman filtering to reduce the influence of noise before the original data is used, so as to obtain the relatively optimal estimated value at the current moment.

9. The method of claim 2, wherein the method comprises the following steps: the acceleration intensity value is the vector sum of the acceleration of the head of the driver on the coordinate axis calculated through an algorithm.

10. The method of claim 2, wherein the method comprises the following steps: the angular velocity strength values are the combined angular velocity values in different directions, and the angular velocities in different directions are integrated through a set algorithm, so that the complexity of angular velocity analysis is reduced.

11. The method of claim 2, wherein the method comprises the following steps: the impact duration is the time of the whole process of the driver's head lowering, and the normal head lowering and the head lowering caused by fatigue napping are distinguished through a threshold value.

Technical Field

Internet of things technology

Background

The technology of the internet of things is one of the technologies for comparing fire and heat in recent years, and the rise of the technology of the internet of things brings a new solution for fatigue detection. A method for detecting fatigue of a Bluetooth headset integrates an audio coding and decoding module, a communication module, a network module, a power supply module, an accelerometer and the like. The method mainly comprises the steps of estimating the head posture of a driver, enabling the head of the driver to generate state change with rapid change due to fatigue, such as rapid head lowering posture caused by dozing, enabling the head to have rapid changes in acceleration, angular velocity and time difference in different directions in the process, resolving the head posture of the driver in real time through a related algorithm, calculating the fatigue state, and uploading information to a mobile terminal in a Bluetooth communication mode.

From the currently retrieved data, the appearance of a bluetooth headset that detects fatigue using head pose estimation has not been seen.

Problems to be solved

The invention designs the Bluetooth headset capable of detecting the fatigue state by estimating the head posture, which effectively reduces traffic accidents caused by fatigue of a driver in the driving process while providing an entertainment function.

Disclosure of Invention

In order to reduce the occurrence of traffic accidents caused by fatigue driving of a driver, the invention designs the Bluetooth headset for detecting the fatigue state in a head posture estimation mode. The general structure of the system is shown in fig. 1.

A method for detecting fatigue of a Bluetooth headset comprises an audio encoding and decoding module, a fatigue detection device, a network module, a power supply module and an alarm.

The Bluetooth headset can make and receive calls and listen to music, and is the same as a common headset. The fatigue detection uses an MPU6050 accelerometer module to estimate the head posture state, and when the head is in a low head state caused by fatigue napping is detected, the driver is determined to be in a fatigue driving state, and an audible and visual alarm is triggered at the moment. The specific flow chart is shown in fig. 2.

After the Bluetooth headset is started, after all sensor modules in the Bluetooth headset are initialized, on one hand, common functions (including call receiving and making, music listening and the like) of the headset can be normally used, and on the other hand, the fatigue detection module detects the head posture state of a driver in real time.

Definition of fatigue napping causing a low head: if the front of the human face is set as an x axis, the direction of the vertex is set as a z axis, and the direction vertical to the x axis and the z axis is set as a y axis. The head-lowering can be considered as the acceleration value of the driver's head in the x-axis and z-axis directions becomes a value other than 0. When the driver falls down due to fatigue napping, the software program can resolve the rapidly changing acceleration and angular velocity, and if the driver accords with the fatigue posture, the alarm gives an audible and visual alarm: the alarm will sound while the alarm light is on.

The fatigue detection device needs to acquire triaxial acceleration data and triaxial angular velocity data of the head posture of the human body. The design uses the MPU6050 module as a sensor for human head pose estimation. The MPU6050 sensor module comprises a three-axis acceleration sensor and a three-axis gyroscope, and the three-axis acceleration sensor can detect and output stress in x, y and z directions; the triaxial gyroscope detects the angular velocity in three directions.

When the MPU6050 six-axis inertia measurement module collects three-axis acceleration data and three-axis angular velocity data, some noise always exists due to interference of external environmental factors or influence of self factors, so that filtering processing is required to be performed before the original data is used to reduce influence of the noise. The kalman filter can estimate the random signal and obtain a relatively optimal estimated value at the current moment, so the kalman filter is selected to deal with the noise problem of the MPU6050 during data acquisition. Kalman filtering is essentially a recursive algorithm that describes linear, discrete and infinite dimensional systems with a state space method, where the pre-estimated optimal equation is:

X(k|k-1)＝A(k,k-1)*X(k-1|k-1)+B(k)*u(k)

the value of X (k | k-1) is the estimated value of k time calculated at time k-1, X (k-1| k-1) is the optimal value at time k-1, A (k, k-1) is the state transition matrix of the system, B (k) is the control weighting matrix, and u (k) is the control signal at time k.

The pre-estimated optimal covariance is expressed as:

P(k|k-1)＝A(k,k-1)*P(k-1|k-1)*A(k,k-1)+Q(k)

Q(k)＝U(k)*U(k)

p (k | k-1) and P (k-1| k-1) are both covariance, the former corresponds to the pre-estimated optimal value X (k | k-1), the latter corresponds to P (k-1| k-1), U (k) is the dynamic noise at time k, and Q (k) is the covariance of the system process at time k.

Calculating a Kalman gain matrix as follows:

k (k) is a Kalman gain value, R (k) is the confidence level of the measurement process to the measurement at the moment k, N (k) is the observation noise at the moment k, and H (k) is an observation matrix.

Then the update estimation method can be expressed as

X(k|k)＝X(k|k-1)+X(k)*(Z(k)-H(k)*X(k|k-1))

P(k|k)＝(I-K(k)*H(k))*P(k|k-1)

Where Z (k) is the measurement at time k and I is the identity matrix.

After the noise is processed through Kalman filtering, the obtained triaxial acceleration data and triaxial angular velocity data are closer to actual values, so that the fatigue state of a driver can be judged more accurately.

The general flow chart of data acquisition for fatigue detection is shown in fig. 3.

When a driver wears the earphone, the front of the face of the human body is set to be an x axis, the direction of the top of the head is set to be a z axis, and the direction vertical to the x axis and the z axis is set to be a y axis. When a driver is in a fatigue state, the head of the driver can automatically lower, and the acceleration and the angular speed in the three directions can be changed to different degrees. The fatigue state of the driver is judged by the following three criteria.

(1) Acceleration intensity value A

If with a_xAnd a_zRespectively representing the acceleration values of the driver's head in the x-axis direction, a_x ²And a_z ²Can be expressed as a change in kinetic energy produced by a change in acceleration of the driver's head in the x-axis direction. The vector sum of the acceleration is selected as a judgment unit of the head movement of the driver, so that the influence caused by uncertainty of the direction of the collected data is eliminated. The acceleration intensity value, a, is set, which can be expressed as,

this a value is close to 0 when the driver is driving normally. If the driver has a head-down phenomenon, the value of A is greater than 0.

(2) Angular velocity intensity value G

If using G_xAnd G_zRepresenting the angular velocity values, G, of the driver's head in the x-axis and z-axis directions, respectively_x ²And G_z ²Expressed as the intensity of the driver's head movements in the x-axis and z-axis directions, respectively. The desired angular velocity is now selected to avoid the complexity of the analysis in different directions, which may be referred to as the angular velocity intensity value G,

the value of G at the time of lowering the head of the driver due to fatigue is different from the value of G at the time of lowering the head in the normal state. It is found through experiments that when the G value is larger than 0.675321rad/s, the driver is judged to have a head drop due to fatigue.

(3) Duration of impact T

In the above formulaThe time at which the head-down starts is indicated,represents the time at which the lowering ends, and T represents the time at which the lowering process takes place entirely. The time for the head-down due to fatigue is generally shorter than the normal head-down time (the normal head-down time is 0.6 seconds).

The acceleration intensity value A and the impact time length T are first judgment conditions, and when the acceleration intensity value A is larger than 0 and the impact time length is smaller than 0.6 second, the suspected fatigue state is judged. Under the condition, if the angular velocity intensity value G is larger than 0.675321rad/s, the driver is judged to be fatigue driving.